Injected Lycoming fuel pressure fluctuation

Jim K

Jim K

Final Approach
PoA Supporter
Joined
Mar 31, 2019
Messages
5,970
Location
CMI
Display Name
Display name:
Richard Digits
On the flight from San Antonio to Corpus Christi, a yellow number on the engine monitor caught my eye. The video below shows what was happening. The engine seemed unaffected & fuel flow was normal. Turned the boost pump on, and the pressure increased, but continued to fluctuate by 10 psi. Switching tanks had no effect.


Posted this to the pa32 group, who mostly seem to think it's normal. I don't think it has done this in the past, but until I dump the engine monitor I can't be certain. I do think I would've noticed a yellow number if it had happened before.

Any ideas? There's a mechanic on the field. I'm not crazy about turning a random mechanic loose on it either, but it's making me nervous.

 
Is this only a problem in cruise? What about idle? How many hours on the engine (fuel pump)?
 
Joe_B1 said:
Bad sensor?
Click to expand...
That's my mechanic at home's opinion. There's a lot of discussion on the internet about the presence/ absence of air in the pressure sensor line

MooneyDriver78 said:
Is this only a problem in cruise? What about idle? How many hours on the engine (fuel pump)?
Click to expand...
First noticed on climbout, continued throughout the flight at various power settings. It was remarkably consistent. Once we were on the ground I was preoccupied with parking and everyone whining about being hungry. I need to go do a ground run and see what it does. I believe the fuel pump was last replaced at engine overhaul, which was about 850 hours and 15ish years ago.
 
Jim K said:
There's a lot of discussion on the internet about the presence/ absence of air in the pressure sensor line
Click to expand...

Come to think of it, we were idling for about 10 minutes waiting for takeoff clearance. Wonder if it got hot enough to vaporize fuel in the line.
 
Jim K said:
There's a lot of discussion on the internet about the presence/ absence of air in the pressure sensor line
Click to expand...

It shouldn't matter. Further, air is compressible and fluid is not. It would actually have a damping effect on the fluctuation.

I don't know what the acquisition rate is on the engine monitor but I suspect it is pretty slow. If you actually had some high speed sensors and acquisition you'd likely see pulsing in the fuel pressure assuming your engine has the typical diaphragm style fuel pump on it. This problem could be something or it could be nothing, do you have a mechanical fuel pressure gauge to reference?
 
mondtster said:
It shouldn't matter. Further, air is compressible and fluid is not. It would actually have a damping effect on the fluctuation.

I don't know what the acquisition rate is on the engine monitor but I suspect it is pretty slow. If you actually had some high speed sensors and acquisition you'd likely see pulsing in the fuel pressure assuming your engine has the typical diaphragm style fuel pump on it. This problem could be something or it could be nothing, do you have a mechanical fuel pressure gauge to reference?
Click to expand...
No...the 930 is certified as primary, and everything but the MP/FF and tach were removed.

It seems to me....by my admittedly elementary understanding of the system.... that if fuel flow is steady and pressure is all over the place, one of them has to be in error. EGTs and performance being stable, it seems the pressure reading must be wrong.
 
Jim K said:
It seems to me....by my admittedly elementary understanding of the system.... that if fuel flow is steady and pressure is all over the place, one of them has to be in error. EGTs and performance being stable, it seems the pressure reading must be wrong.
Click to expand...

There may be no error at all; it may just be a fluke that the timing of the pressure pulses happened to fall at the right time so you see it in the measurement. If you put a mechanical fuel pressure gauge with no snubber on the fuel pump output you'd see the pulsing from a diaphragm pump. The local flight school's Arrow has done it for years and 2+ engines run through it. The acquisition rate and transducer speed on your engine is likely slow enough you typically can't see it or it gets "averaged" to numb it a bit.

But you may also have a flaky transducer. I'd check the fuel pressure with a mechanical gauge and see what it looks like. Whether you wait until you get home or do it on the spot is the decision you'd need to make.
 
my vote, with steady fuel flow and no uncommanded power surge loss during the episode, that's a finnicky pressure sensor. I'd fly home. But you're not me and I'm not you, ultimately your call.
 
So I dumped my engine monitor log today. I actually found the flight where the pressure started fluctuating. Up until about a month ago it ran a steady 22-23 psi with very little variation. In the middle of one otherwise completely unremarkable flight, it started bouncing between 20-26. On the TX trip it bounced from 15-30. Seems like the mean is still 23 psi. I'm convinced it's an indication issue. Fuel flow and EGT's still rock solid. None of the other logged parameters changed either, so I don't think it's likely to be a ground issue.

I did engage the mechanic in Corpus Christi to help me check both fuel screens, which were completely spotless, and look for leaks, which we never found, so we went ahead and flew home uneventfully.

Called JPI and they were no help whatsoever. "sounds like a wiring problem". Thanks. They claim the sensors never fail :dunno: The sensor is made in Germany, so I'm sure it's stupid expensive. I'm going to try bleeding the line to the sensor. If that doesn't help my local mechanic is scheduled to look at it Monday. We'll see if a mechanical fuel gauge shows the same fluctuation.

Anybody want to takes bets whether the fuel pump or the sensor is cheaper? I'm sure I'll end up replacing both and it'll be the second one that fixes it.

Is it possible one of the reed valves in the mechanical pump is sticking/leaking and the pressure is actually varying that much? I would think the fuel servo would have a hard time keeping up with that much variation, but maybe because the average pressure is still okay by the time it gets to the injectors the pulses have evened out?
 
Resolution!

Went back to the airport today to finish up my oil change. Bled the line to the fuel pressure sensor... no change. Then i opened up the line and let the fuel OUT. Fired her up and the fuel pressure was rock solid like it used to be. Apparently the air "spring" trapped in the line is what smooths out the pump pulses.
 
Jim K said:
Resolution!

Went back to the airport today to finish up my oil change. Bled the line to the fuel pressure sensor... no change. Then i opened up the line and let the fuel OUT. Fired her up and the fuel pressure was rock solid like it used to be. Apparently the air "spring" trapped in the line is what smooths out the pump pulses.
Click to expand...

Would you expound upon what you’re saying here, I’m not quite following. I have the EI engine monitors and my fuel pressures are all over the place (twin), similar to yours. We’ve also tried bleeding the line to the sensor with no resolution, so I just ignore them for now.
 
Banjo33 said:
Would you expound upon what you’re saying here, I’m not quite following. I have the EI engine monitors and my fuel pressures are all over the place (twin), similar to yours. We’ve also tried bleeding the line to the sensor with no resolution, so I just ignore them for now.
Click to expand...

Air is compressible, fluid is not. Leaving some air trapped in the line going to the pressure transducer will "numb" the pulsation. You could also try adding a snubber to the line.

Here's an example.

https://www.amazon.com/Auto-Meter-3...t=&hvlocphy=9018109&hvtargid=pla-761192406472
 
Banjo33 said:
Would you expound upon what you’re saying here, I’m not quite following. I have the EI engine monitors and my fuel pressures are all over the place (twin), similar to yours. We’ve also tried bleeding the line to the sensor with no resolution, so I just ignore them for now.
Click to expand...
What @mondtster said...i basically created an air spring in the line to dampen the pulsations of the pump.

I tried bleeding all the air out at first, but that didn't change the fluctuations.

I next opened the line at the bulkhead fitting on the firewall, which is high, and a union that is about 6"lower, but still about 8" above the fuel servo. I let the fuel drain out, and then closed the lines, trapping some air. Because this is all higher than the pump, and deadheads at the pressure sensor, the air has no where to go. After I did that, my fuel pressure reading is rock solid like it used to be. I assume the air will work its way out again eventually, but we'll just have to see how long it takes.

Right? Wrong? I don't know, but it solved the issue, at least for the 5 hours or so since it was done.
 
This reminds me of a fix I used to implement on the pneumatic controls of large (like over 100,000 CFM large) air handlers with variable inlet vanes and troublesome hysteresis issues. The fans were in the 125-150 HP range, and if the inlet vane hunting got out of control, the excessive static pressure swings could actually collapse the air handler.

I would make an air tank with capped length of 4" PVC pipe, and connect it to the inlet feed on the static pressure sensing line of the pneumatic controller with a tee. I would put the tank on the back of an I-beam or other concealed space. Problem solved!
 
Banjo33 said:
Is this a legitimate snubber or does anyone know where to find an aviation-specific variant? Another mechanic agreed that a snubber is likely to solve my issue, but his thought was that it actually goes in the fuel line (inside the fitting), but didn’t know where to find one. Spruce search came up empty.
Click to expand...

That was just a 2 second search to provide an example. The approach your mechanic is talking about is to use a restrictor in the fitting to provide a similar service. Lycoming uses restricted fittings in some applications, so buying one of those is likely the path of least resistance. I've also made restricted fittings using drilled set screws threaded into flare fittings but some mechanics might get queasy about that. Here's a restricted fitting from Spruce. Again, it is a 2 second search so there might be something else that better fits your needs.

https://www.aircraftspruce.com/catalog/eppages/oilfuelflow05-11908.php?clickkey=9212
 
Thank you for that! Some of these things are so new to me, so all I have to go off of is terminology. I couldn’t find any reference to a “snubber” in the Lycoming or TwinCo parts manuals. This helps a ton!
 
You must log in or register to reply here.
Back
Top